| dc.description.abstract | Augmented-reality Sandbox allows users to interact with 3-dimensional simulations of a river landscape. In geoscience lessons, it may facilitate the understanding of dynamic processes that consist of complex action layer(s) that are difficult to understand and structure layer(s). Students’ actions play crucial role when learning through Sandbox. Hence, a quasi-experimental design was used to determine how students’ actions, i.e. gestures, when using Sandbox compared to those who used traditional classroom resources. 29 students were divided into Sandbox group (who used sandbox as resource during learning activity) and Paper group (who used flip-chart paper and pen as resource). After learning activity, students had to explain one of the dynamic processes on the resource they had used. The explanations were recorded for speech-gesture analysis. On average, there was no significant difference in numbers of structure or action gestures produced by two groups. Qualitative analysis of videos and pre- and post-tests did not reveal any difference between groups’ understanding of dynamic processes. It was found that Paper group used more deictic (pointing) gestures (t(23) = 2.66, p = .014, d = 1.06). Difference in number of iconic (semantic) gestures was not significant, but a medium effect size was observed (d = 0.66). Further qualitative analysis suggested that Paper group often used pointing gestures to refer to structures and locations. In contrast, Sandbox group often used iconic gestures that represented shapes to refer to same structures and locations. This effect can be attributed to the environment sandbox creates which allows users to perceive structures and locations on landscape in various ways. Hence, sandbox makes more types gestures available to students when communicating. | |
